Chapter 1
Induced Technological Change and
the Environment: An Introduction
Arnulf Grübler, Nebojsa Nakicenovic, and William D. Nordhaus
Technology is a double-edged sword for the environment: it can amplify as well as alleviate the impacts of human activities. The role of technology was largely ignored in the first round of global modeling efforts in the early 1970s, but it has recently moved to the forefront of both science and policy in addressing climate change. The long time span involved in climate change—from decades to centuries—puts technological change “in the driver's seat,” because over long time periods technological systems are highly malleable, whereas in the short run they are largely inflexible.
A brief look at the history of technology illustrates the enormous scope for change: a century ago the industrialized countries relied on a coal-fueled steam-engine economy. Current technologies such as instantaneous global electronic communication or heavier-than-air flying machines were at most fantasies in the minds of science fiction writers. They were certainly not a source of inspiration for individuals in the newly founded research and development (R&D) laboratories. While it is easy to describe the enormous changes in technological hardware, software, and “orgware” that have characterized the past, and while it is comparatively easy to speculate about possible future developments, it is much harder to discern the factors that have caused all these changes. Why were certain technological options pursued while others were ignored? Why did some technologies gain widespread social acceptance and diffusion while others never moved beyond the status of a technological curiosity? And finally, what role was played by institutions and policies in triggering and promoting, or in obstructing and slowing down, change?
New and improved technologies do not “fall from heaven like autumn leaves.” But do we have sufficient scientific knowledge about the sources and management of innovation to properly inform the policymaking process that affects technology-dependent domains such as energy or agriculture and their interactions with the environment? In light of current knowledge, we think the answer is, Not yet. We have theories and we have data, but we continue to lack a comprehensive conceptual framework for integration—a framework that can do justice to the “maze of ingenuity” surrounding technological innovation and the associated uncertainties, to the complexity of factors that govern the incessant modifications and improvements of existing techniques, and to the diverse economic, social, and institutional factors driving their diffusion. Finally, we must recognize the intrinsic challenge of heterogeneity in technologies and agents for both innovation and diffusion.
A series of three workshops convened at the International Institute for Applied Systems Analysis (IIASA) in Laxenburg, Austria, from 1997 to 1999 focused on induced technological change (ITC), addressing the conceptual, empirical, and modeling challenges that an ITC perspective entails. By reviewing the state of the art of ITC theory, empirical case studies, and novel methodological approaches of modeling ITC, the workshops not only served to take stock, but also to generate new ideas and to frame a future research agenda. The current volume reports on the fruits of this collective endeavor. We wish to emphasize the collective nature of this exercise, to which all participants provided vital intellectual inputs. This volume necessarily contains only a limited number of contributions, which have been carefully chosen, reviewed, revised, and edited for the consideration of a wider readership.
Foremost, therefore, our sincere thanks go to the authors, the anonymous reviewers, and the workshop participants for their intellectual contributions and open minds, which enabled all of us to learn from one another and to report on the knowledge gained—as well as the knowledge gaps remaining. The contributors to this volume have been particularly generous with their time and efforts, not only in preparing draft manuscripts but also in revising them substantially in light of many fruitful discussions and the customary peer review process. Some contributions have been published in the meantime in a variety of scholarly journals. They are included here for the sake of completeness. The publishers—acknowledged in the relevant chapters—were kind enough to grant permission to reprint the articles so that the intellectual integrity of the contributions could be maintained. Transforming these individual contributions into a coherent volume is far from an easy task. Special thanks therefore go to the IIASA Publications Department for dealing with all technical production aspects so competently and expeditiously. Finally, we wish to acknowledge the financial support from IIASA, the United States National Science Foundation (which provided the resources through the Yale–NBER–IIASA program on international environmental economics), and the Austrian Federal Ministry for Education, Science and Culture. None of these institutions is responsible for the views expressed in this volume.
The individual chapters in this book are grouped into three main parts and are summarized in more detail below to guide the reader through this volume. Part One, including this overview (Chapter 1), provides historical background on the ITC debate (Vernon Ruttan, Chapter 2) and frames the issue from both a historical (Joel Mokyr, Chapter 3) and an international (Robert Evenson, Chapter 4) perspective. Part Two gives empirical insights into patterns of technological change either in response to price signals (Richard Newell and colleagues, Chapter 5) or technology policy (Chihiro Watanabe and colleagues, Chapter 6), or from a historical, evolutionary point of view (Nebojsa Nakicenovic, Chapter 7). Part Three presents recent ITC modeling approaches, embracing both “top-down” (William Nordhaus, Chapter 8; Lawrence Goulder and Koshy Mathai, Chapter 9) and “bottom-up” perspectives (Andrii Gritsevskyi and Nebojsa Nakicenovic, Chapter 10; Arnulf Grübler and Andrii Gritsevskyi, Chapter 11). It also provides an extensive literature review as well as reflections on the modeling research agenda ahead (Leon Clarke and John Weyant, Chapter 12). The final chapter (Vernon Ruttan, Chapter 13) moves the debate forward by extending the technological dimension into the realm of institutional innovations, a critical area when considering inducement mechanisms of technological change beyond the traditional dichotomy of “supply-push” versus “demand-pull” paradigms of research and policy.
Chapter 2, by Vernon Ruttan, reviews the historical roots of ITC as well as recent research streams. The chapter reviews the historical “demand-pull/supply-push” controversy and discusses the first ITC formulations in both their microeconomic and growth-theoretic foundations. The chapter then discusses two more recent research streams, namely, evolutionary theory and concepts of “path dependency” (often referred to also as technological “lock-in” phenomena). Of particular interest is Ruttan's careful assessment of the relative strengths and weaknesses of each research stream and his discussion of the need for a constructive dialogue between theory and empirical data. Ruttan reaches the conclusion that each individual research tradition—while having generated substantial insight into the generation and choice of new technology—is showing diminishing returns, but that, significantly, the different models of technological change (induced, evolutionary, path-dependent) can be seen as elements of a more general theory. The chapter makes concrete and useful suggestions toward the integration of different theoretical ITC streams. It concludes with a call for further integration of the insights gained from the theoretical and empirical research conducted within an ITC perspective and endogenous economic growth theory in order to gain new insights into the relationships between human capital, scale, and trade in the process of economic growth and development.
Chapter 3, by Joel Mokyr, provides an evolutionary interpretation of technological innovation in medical history. The chapter puts knowledge at the center of the debate. The history of medical technologies and techniques is indeed a most appropriate case study: nowhere is “demand-pull” as powerful as in the human desire to lead a long and healthy life. Yet, for all that demand-pull, improved medical technology will not emerge without adequate and improved knowledge. Quite appropriately, Mokyr refers to the “necessity is the mother of invention” theory of technological change as both a cliché and a historical fallacy. By developing a theoretical model of useful knowledge, Mokyr describes ITC as emerging from changes in knowledge as well as from changes in the selection environment of technological traits that can be realized with that knowledge. The evolution of knowledge in turn is seen as the net historical result of variation versus selective retention. From such a perspective, the chapter illustrates the limits of our capacity to steer or even to predict the evolution of new knowledge. Clearly, available resources and institutions are some of the factors that determine how conducive the existing knowledge base and society at large are to its expansion. But the uncertainties in ultimate outcomes remain enormous, with surprise and disappointment invariably accompanying the quest for new (technological) knowledge.
Chapter 4, by Robert E. Evenson, concludes Part One of this volume by explicitly addressing the international dimension of technological innovation and diffusion and the resulting impacts on productivity levels. Contrary to earlier expectations, empirical evidence suggests that productivity levels in developing countries have converged in the agricultural rather than the industrial sector. Evenson discusses two alternative mechanisms for productivity convergence to explain the puzzle: a mimicry mechanism and an induced adaptive invention mechanism with international invention recharge. The latter model requires both an endogenous capacity for innovation/adaptation in developing countries as well as a continuous “recharge” mechanism, by which innovation possibilities are replenished, for example, in the form of germplasm for the breeding of new plant varieties. This recharge mechanism, on which Evenson has pioneered research, highlights the importance of continuous investments in the knowledge base from which innovations can emerge. Evenson's tests of the model with empirical data (germplasm and diffusion of high-yield varieties in agriculture, international patent data in industry) clearly support the model. What seems even more important, however, is that the model of induced adaptive innovation with international recharge has been institutionally internalized in agriculture—leading to productivity convergence—whereas in industry no similar broad-based institutions facilitating local adaptation and assimilative capacity of innovations as well as an international recharge mechanism are available in developing countries. From that perspective, it is less surprising to see that convergence in industrial productivity levels is confined to the Organisation for Economic Co-operation and Development “club” and a small group of newly industrializing developing countries.
Chapter 5, by Richard G. Newell, Adam B. Jaffe, and Robert N. Stavins, opens Part Two of this volume by dealing with empirical analyses and findings from an ITC perspective. Using the novel approach of “characteristics transformation surfaces” for three consumer durables over several decades, they test a Hicksian inducement mechanism: how changing energy prices have affected the characteristics along the technological “possibility” frontier (otherwise frequently referred to as “best available technology,” or BAT) as well as in the composition of product models available on the market (product substitution). Their analysis also includes the impacts of government regulations, including labeling and mandatory efficiency standards. The authors find that the overall rate of innovation was independent of prices and government regulation, but that the direction of technological change was responsive to price changes in a number of instances. Changes in energy prices in particular induced significant changes in the subset of models offered on the market, and this responsiveness increased substantially in the presence of labeling requirements. Finally, mandatory efficiency standards also had a significant impact on the average energy efficiency of the product menu available to consumers. Nonetheless, a sizable portion of energy-efficiency improvements was still found to be unrelated to changes in energy prices or regulation, pointing to the multitude of factors influencing the pace and direction of technological change.
Chapter 6, by Chihiro Watanabe, Charla Griffy-Brown, Bing Zhu, and Akira Nagamatsu, offers a unique glimpse into the workings of the Japanese system of innovation by examining in detail the spillover effects and positive feedback mechanisms at work in the development of photovoltaic (PV) technology. By drawing on a unique data set, the authors are not only able to shed light on knowledge (R&D) spillover effects among leading Japanese PV firms (a central area of concern regarding knowledge externalities inherent in technological innovation), but are able to relate it to the “virtuous cycle,” most prominently reflected in the tremendous cost decline (of well over a factor of 40) in the economics of PVs over the past few decades. As such, the chapter breaks new ground in our understanding of the microeconomic foundations of technological “learning” phenomena, combining both supply- and demand-led inducement factors. Evidently, the ITC perspective offered in this chapter reports on a rather unique experience. The inducement effect of public R&D funding (by MITI, the Japanese Ministry for Trade and Industry) on private R&D, and its effect on improvements in PV technology, relies both on the specific coordination mechanisms that characterize the Japanese system of innovation as well as on a variety of demand-side factors. The latter include both a “moving target” of guaranteed PV purchase prices as well as the familiar impact of declining costs on market growth for a new technology. Perhaps the most important message to retain from this detailed analysis is the extreme importance of taking a long view in technology policy and the existence of an appropriate institutional setting.
Chapter 7, by Nebojsa Nakicenovic, explores the nature of the relationship between technological change, the costs and performance of new technologies, and their effect on reducing carbon dioxide (CO2) emissions. The chapter shows that an important part of the secular decline of energy and carbon intensities is the result of technological change. Technologies that are more energy efficient have replaced less efficient ones, and technologies that are less carbon intensive have replaced those that are more carbon intensive. Further, it is argued that an important component of the dynamics of technological substitution is a cumulative process of learning by doing and that timely investment in new technologies with lower CO2 emissions is a cost-effective strategy for reducing global carbon emissions. A number of implications are considered with reference to emission mitigation strategies. One is that there may be great leverage in policies that accelerate the accumulation of experience in new technologies with lower environmental impacts, for example, through early adoption and development of special niche markets. This leverage can be important, particularly if these policies can minimize the “deadweight” loss to society associated with the foregone exploitation of cheaper fossil fuels and possible reductions of research, development, and demonstration (RD&D) efforts in other parts of the economy.
Chapter 8, by William D. Nordhaus, opens Part Three of this volume with an analysis of the relative importance of ITC in the context of climate-change policy. Nordhaus notes that most studies of environmental and climate-change policy have ignored the thorny issue of induced innovation by assuming that technology is exogenous. He analyzes the impact of induced innovation by developing a model of induced innovation and incorporating this in an updated version of the globally aggregated DICE model called the R&DICE model. The approach specifies a model of induced innovation in which the stock of carbon-energy technological knowledge responds both to technological developments outside the carbon-energy sector and to profit-oriented R&D within it. The specification and parameters of the induced-innovation equations are determined from the extensive work available on the returns to R&D. The principal conclusion of the study is that, over the next century, induced innovation is likely to be a less powerful factor in implementing climate-change policies than substitution. The reductions in CO2 concentrations and in global mean temperature due to induced innovation are estimated to be approximately one-half those due to substitution. If confirmed by further research, this study suggests that subsidizing R&D or energy technology is unlikely to be a fruitful approach to solving the climate-change problem.
Chapter 9, by Lawrence H. Goulder, and Koshy Mathai, examines optimal CO2 abatement profiles under various model specifications of ITC. The model considers two specifications for technological progress: one relying on an R&D-based regime of knowledge accumulation, and the other relying on a learning-by-doing formulation. These two formulations are then tested for their impact on both the timing and extent of emission abatement as well as on optimal carbon tax levels under both a cost-effectiveness and a benefit-cost criterion using a by now classic formulation first proposed by Nordhaus where abatement costs are contrasted with a nonlinear climate-change damage function. The particular appeal of the Goulder/Mathai approach lies, first, in the analytical solutions to the problem formulation, and, second, in the extensive sensitivity analyses performed. As expected, the presence (or absence) of ITC has its most significant impacts on lowering (or increasing) optimal carbon tax profiles, assuming of course that the shape and parameters of the climate damage function are known with certainty. Conversely, the impacts of ITC on the timing of emission abatement are more ambiguous and depend on the assumed regime of knowledge accumulation (R&D versus learning by doing). The Goulder/Mathai model and the numerical simulations discussed in the chapter provide valuable guidance for future research, proposing, first, a more detailed examination of the microeconomic foundations and complexities of different forms of knowledge accumulation, and, second, representing the numerous uncertainties involved in knowledge accumulation and in the abatement cost and climate damage functions.
Chapter 10, by Andrii Gritsevskyi and Nebojsa Nakicenovic, presents a new method for modeling induced technological learning and uncertainty in energy systems. Three related features are introduced simultaneously: (1) increasing returns to scale for the costs of new technologies; (2) clusters of linked technologies that induce learning “spillovers” depending on their technological “proximity,” in addition to the technology relations through the structure (and connections) of the energy system; and (3) uncertain costs of all technologies and energy sources. One of the results of the analysis is that there is a large diversity across alternative energy technology strategies. The strategies are path dependent, and it is not possible to choose a priori the “optimal” direction of energy systems development. Another result of the analysis is that endogenous technology learning with uncertainty and spillover effects will have the greatest impact on the emerging structures of energy system during the first few decades of the next century. Finally, the results imply that fundamentally different future energy system structures might be reachable with similar overall costs. Thus, future energy systems with low CO2 emissions need not ...